Objective—To compare the findings of light microscopic evaluation of routine unstained wet-mounted preparations and air-dried, modified Wright-stained preparations of urine sediment with results of quantitative aerobic bacteriologic culture of urine.
Design—Masked prospective study.
Sample Population—459 urine samples collected by cystocentesis from 441 dogs.
Procedure—Urinalyses and quantitative bacteriologic cultures of urine were performed. Unstained wetmounted preparations and air-dried, modified Wrightstained urine sediment preparations were examined by light microscopy for the presence of bacteria.
Results—Compared with results of quantitative bacteriologic culture, routine unstained preparations and modified Wright-stained preparations had sensitivities of 82.4% and 93.2%, specificities of 76.4% and 99.0%, positive predictive values of 40.1% and 94.5%, negative predictive values of 95.8% and 98.7%, and test efficiencies of 77.3% and 98.0%, respectively. Compared with 74 samples that yielded growth on bacteriologic culture, the routine unstained method had concordance and misclassification rates of 39.2% and 60.8%, respectively, whereas the Wright-stained method had concordance and misclassification rates of 78.4% and 21.6%, respectively. Significant associations between each of occult blood in urine, pyuria, female sex, and lower urine specific gravity with bacteriuria detected by Wright-stained sediment examination and quantitative bacteriologic culture of urine were identified.
Conclusions and Clinical Relevance—Examination of modified Wright-stained preparations of urine sediment appeared to be a rapid, cost effective method that significantly improved the sensitivity, specificity, positive predictive value, and test efficiency of light microscopic detection of bacteriuria, compared with that of the routine unstained method. (J Am Vet Med Assoc 2004; 224:1282–1289)